Prof. F. J. Bell on the Lagan idse. 135 



forms ; and they may be summed up in the following 

 Table : 



Extent of 



B '*\%»J - *^\%\ T" 



partition. 



x ercent 



3 



27-27 



4 



3(5-36 



8-5 



44-7 







42-8 



23 



M 



27 



68-4 



Species. Radius. 



L. Putnami ] 1 



M P. decagonalis " (younp) . ] 1 



" L. Peronii " 19 



L. depressum 21 



u P. deca<ronalis v 41 



" P. deeagonalis " 4(3 



It is clear enough that these figures do not support the 

 doctrine of a marked difference between the species of Laga- 

 num and Peronella in the extent of the development of the 

 partitions ; but they do point in a most significant manner to 

 the apparent relation between the size of the test and the 

 development of the internal supporting walls ; and we are led 

 to see without surprise that a species placed by one naturalist 

 with those in which the u sinus cecaux du bard " are " limites 

 a une zone etroite," may be set by his son among those in 

 which the partitions are said to ramify and extend some 

 distance inwards (L. decagonum) . 



In fine, we are irresistibly led to the conclusion that these 

 partitions present us with variations which are largely due to 

 growth, and, for the rest, available only as marks of specific 

 differentiation. 



The fact that in the technical definition of the genus 

 Laganum A. Agassiz says " five genital openings/ 7 and in 

 that of the subgenus Peronella u four genital openings,' r 

 would lead the commencing or unwary student to suppose 

 that that author attached some weight to the difference in 

 number. As a matter of fact, however, he (and, as it would 

 seem, with perfect justice) unites, under the head of P. deca- 

 gonaltSj forms with four and forms with five pores. 



3. General Results of the Investigation. — The evidence 

 adduced leads to the union of Peronella (A. Ag.) with Laga- 

 num. In the examination of the question of the character 

 of the systematic relations of the species of the Laganidae, we 

 have been compelled to enter with as much detail into the 

 views of earlier naturalists as into the lessons to be learnt 

 from structural characters. We have found that, driven from 

 point to point, they had come to see that in the characters of 

 the pores, in the form of the partitions, the only two points 

 of difference were to be expected. Some forty years ago the 

 elaborate studies of Louis Agassiz convinced him that the 

 variations in the number of the pores could not be brought 

 into association with any other variations that could be re- 



